smids_5x_deit_tiny_rms_00001_fold4

This model is a fine-tuned version of facebook/deit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.4112
• Accuracy: 0.8783

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2498	1.0	375	0.3750	0.8583
0.2126	2.0	750	0.3946	0.8617
0.0815	3.0	1125	0.3928	0.8817
0.1183	4.0	1500	0.4272	0.8733
0.1029	5.0	1875	0.5782	0.8833
0.0245	6.0	2250	0.6426	0.8867
0.0551	7.0	2625	0.8096	0.8733
0.0319	8.0	3000	0.8011	0.8733
0.0533	9.0	3375	0.8429	0.875
0.0056	10.0	3750	0.9672	0.8617
0.0136	11.0	4125	1.0120	0.8667
0.0031	12.0	4500	0.9881	0.87
0.0176	13.0	4875	1.1184	0.8767
0.0127	14.0	5250	1.1325	0.8583
0.0003	15.0	5625	1.2848	0.8683
0.0058	16.0	6000	1.1232	0.87
0.0002	17.0	6375	1.0571	0.8817
0.0421	18.0	6750	1.2079	0.8717
0.0004	19.0	7125	1.2753	0.87
0.0001	20.0	7500	1.3783	0.86
0.0	21.0	7875	1.3177	0.865
0.002	22.0	8250	1.3637	0.8633
0.0002	23.0	8625	1.4459	0.87
0.0005	24.0	9000	1.2813	0.875
0.0	25.0	9375	1.2487	0.88
0.0	26.0	9750	1.2405	0.875
0.0008	27.0	10125	1.3345	0.885
0.0001	28.0	10500	1.5106	0.865
0.0	29.0	10875	1.2765	0.8733
0.0	30.0	11250	1.2626	0.875
0.0332	31.0	11625	1.3653	0.8667
0.0	32.0	12000	1.3469	0.8683
0.0	33.0	12375	1.2524	0.8817
0.0	34.0	12750	1.2947	0.8767
0.0	35.0	13125	1.2962	0.8733
0.0	36.0	13500	1.3559	0.8783
0.0	37.0	13875	1.3878	0.8817
0.0033	38.0	14250	1.3553	0.8767
0.0	39.0	14625	1.4121	0.875
0.0	40.0	15000	1.4174	0.875
0.0	41.0	15375	1.4132	0.875
0.0	42.0	15750	1.4182	0.8767
0.0	43.0	16125	1.4186	0.8767
0.0	44.0	16500	1.4200	0.8767
0.0	45.0	16875	1.4125	0.8783
0.0	46.0	17250	1.4134	0.88
0.0	47.0	17625	1.4114	0.8783
0.0	48.0	18000	1.4108	0.8783
0.0	49.0	18375	1.4113	0.8783
0.0	50.0	18750	1.4112	0.8783

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.1+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2